As stated above, our study focuses on the inference (i.e. deployment) stage in the model life cycle,
And this is why.
The model cards for Stable Diffusion 1.5 and 2.1 estimate the CO2 emissions as 11.25 tons and 12 tons for training. XL lacks the info.
A transatlantic flight (round-trip) is about 1 ton per pax. So, while every little bit helps, ML is not where you can make the big gains in lowering emissions.
If I remember correctly, SDXL is a heavily modified SD2.1, so the numbers might be similar.
Bullshit. It has to be way more than that.
And this is why.
The model cards for Stable Diffusion 1.5 and 2.1 estimate the CO2 emissions as 11.25 tons and 12 tons for training. XL lacks the info.
A transatlantic flight (round-trip) is about 1 ton per pax. So, while every little bit helps, ML is not where you can make the big gains in lowering emissions.
If I remember correctly, SDXL is a heavily modified SD2.1, so the numbers might be similar.